Abstract: Clonal diseases of large granular lymphocytes (LGL) are characterized by lymphocytosis and T cell-mediated cytopenias of the myeloid compartment. Several syndromes of T cell-mediated bone marrow failure have similar pathology including aplastic anemia (AA), paroxysmal nocturnal hemaglobinuria (PNH), and a subset of patients with Myelodysplastic Syndrome (MDS). All of these syndromes are characterized by clinical improvement with immunosuppressive therapy (IST) such as low-dose methotrexate (MTX), low- dose cyclophosphamide (CY) or cyclosporine A (CyA) and, in the case of aplastic anemia and MDS, T cell depletion with anti-thymocyte globulin (ATG). The broad long-term goal of this proposal is to improve the diagnosis and treatment of patients with LGL leukemia and these other bone marrow failure diseases. In patients with these T-cell mediated bone marrow failure syndromes, it is not clear whether the T lymphocytes become activated locally in the bone marrow against specific bone marrow antigens, or within the peripheral lymphoid system and then home to the bone marrow. The prevailing idea during the last funding period was that an antigen, presumably a common antigen on myeloid progenitors was responsible for suppression of myelopoiesis. This was thought to be due to the direct interaction between this putative myeloid-specific antigen and the expanded antigen-specific T cell clone. In the Progress Report, our data strongly suggests that a multi-step model of autoimmunity for this disease process based on work during the last funding period. We show in preliminary results that the lymph node homing receptor L-selectin (CD62L) is dramatically lost from CD8+ T cells in LGL leukemia patients (manuscript published in Blood 2009). The main hypothesis to be addressed in this proposal is that altered bone marrow homing is critical for LGL leukemia pathogenesis. We hypothesize that acute activation in the primary lymphoid system leads to cleavage and shedding of CD62L by the Tumor Necrosis Factor- Converting Enzyme (TACE), which also known as ADAM-17. This matrix metalloproteinase (MMP) cleaves not only CD62L to control the egress of T cells from the lymph node but it is also necessary for the activity of several inflammatory cytokines with known importance in LGL leukemia and other autoimmune diseases. In Specific Aim 1, we will confirm our hypothesis that loss of CD62L expression is mediated primarily by ectodomain shedding by the ADAM-17 matrix metalloproteinase. Because CD62L is a lymphoid homing receptor, we hypothesize that loss of this receptor then allows these activated T cells to exit the lymph node but loss of CD62L alone is not expected to trigger migration and colonization in the bone marrow where these cells suppress hematopoiesis. The focus of Specific Aim 2 is to determine the mechanism of bone marrow homing by T cells in LGL leukemia using an in vitro system. Homing and migration are complex events that are optimally monitored in vivo and there is no mouse model of LGL leukemia or bone marrow failure disease currently available. Moreover, the regulation of homing to the bone marrow even under normal conditions is incompletely understood. Since the antigen that activates T cells in LGL leukemia is unknown, we will use a well defined transgenic mouse model to study important aspects of bone marrow homing and hemosuppression by antigen-specific T cells. This will allow us to test the importance of CD62L, VLA-4, and ADAM-17 in vivo. Additionally, the efficacy of pharmacological inhibitors to block homing to the bone marrow will be tested to provide critical pre-clinical data for application to clinical studies in patients in the future.